Factor analysis apparatus, factor analysis method, and non-transitory storage medium

A factor analysis apparatus that analyzes a relationship between a target event that is a target of factor analysis and an assumed factor that is assumed to be a factor of the target event, the factor analysis apparatus comprising:

• a memory storing a program configured to at least:
• calculate a first feature of a name of a data item included in provided time-series data,
• extract a term that is similar to the data item from the memory on the basis of the first feature,
• calculate a second feature of a name of the assumed factor,
• extract a term that is similar to the assumed factor from a storage on the basis of the second feature,
• calculate, on the basis of a first term group extracted based on the first feature and a second term group extracted based on the second feature, a degree of name similarity between names of the data item included in the provided time-series data and the assumed factor,
• calculate a first degree of influence indicating a degree of influence of the data item on the target event on the basis of time-series data of the data item and time-series data of the target event,
• calculate a second degree of influence indicating a degree of influence of the assumed factor on the target event on the basis of the degree of name similarity and the first degree of influence,
• calculate a correlation value between estimated time-series data of the target event and actual time-series data of the target event, and
• output an image comprising the target event and the assumed factor, wherein a displayed size of the assumed factor is adjusted to correspond to a magnitude of the second degree of influence;
• and
• a processor configured to execute the program from the memory.

A factor analysis apparatus that analyzes a relationship between a target event that is a target of factor analysis and an assumed factor that is assumed to be a factor of the target event, the factor analysis apparatus comprising:

• a memory storing a program configured to at least:
• calculate a first feature of a name of a data item included in provided time-series data,
• extract a term that is similar to the data item from the memory on the basis of the first feature,
• calculate a second feature of a name of the assumed factor,
• extract a term that is similar to the assumed factor from a storage on the basis of the second feature,
• calculate, on the basis of a first term group extracted based on the first feature and a second term group extracted based on the second feature, a degree of name similarity between names of the data item included in the provided time-series data and the assumed factor,
• calculate a first degree of influence indicating a degree of influence of the data item on the target event on the basis of time-series data of the data item and time-series data of the target event,
• calculate a second degree of influence indicating a degree of influence of the assumed factor on the target event on the basis of the degree of name similarity and the first degree of influence,
• acquire time-series data,
• output an image comprising the target event and the assumed factor, wherein a displayed size of the assumed factor is adjusted to correspond to a magnitude of the second degree of influence;
• and
• a processor configured to execute the program from the memory.

A factor analysis method for analyzing a relationship between a target event that is a target of factor analysis and an assumed factor that is assumed to be a factor of the target event, the factor analysis method comprising:

• calculating a first feature of a name of a data item included in provided time-series data;
• extracting a term that is similar to the data item from the memory on the basis of the first feature;
• calculating a second feature of a name of the assumed factor;
• extracting a term that is similar to the assumed factor from a storage on the basis of the second feature;
• calculating, on the basis of a first term group extracted based on the first feature and a second term group extracted based on the second feature, a degree of similarity between the data item included in the provided time-series data and the assumed factor;
• calculating a first degree of influence indicating a degree of influence of the data item on the target event on the basis of time-series data of the data item and time-series data of the target event;
• calculating a second degree of influence indicating a degree of influence of the assumed factor on the target event on the basis of the degree of name similarity and the first degree of influence;
• and
• outputting an image comprising the target event and the assumed factor, wherein a displayed size of the assumed factor is adjusted to correspond to a magnitude of the second degree of influence.

A non-transitory storage medium storing a program for analyzing a relationship between a target event that is a target of factor analysis and an assumed factor that is assumed to be a factor of the target event, the program being for causing a computer to perform:

• calculating a first feature of a name of a data item included in provided time-series data;
• extracting a term that is similar to the data item from the memory on the basis of the first feature;
• calculating a second feature of a name of the assumed factor;
• extracting a term that is similar to the assumed factor from a storage on the basis of the second feature;
• calculating, on the basis of a first term group extracted based on the first feature and a second term group extracted based on the second feature, a degree of similarity between the data item included in the provided time-series data and the assumed factor;
• calculating a first degree of influence indicating a degree of influence of the data item on the target event on the basis of time-series data of the data item and time-series data of the target event;
• calculating a second degree of influence indicating a degree of influence of the assumed factor on the target event on the basis of the degree of name similarity and the first degree of influence;
• and
• outputting an image comprising the target event and the assumed factor, wherein a displayed size of the assumed factor is adjusted to correspond to a magnitude of the second degree of influence.